文献类型：期刊文献

英文题名：Compact real-valued teaching-learning based optimization with the applications to neural network training

作者：Yang, Zhile[1];Li, Kang[2];Guo, Yuanjun[1];Ma, Haiping[3];Zheng, Min[4]

机构：[1]Chinese Acad Sci, Shenzhen Inst Adv Technol, Shenzhen 518055, Guangdong, Peoples R China;[2]Univ Leeds, Sch Elect & Elect Engn, Leeds LS2 9JT, W Yorkshire, England;[3]Shaoxing Univ, Dept Elect Engn, Shaoxing, Peoples R China;[4]Shanghai Univ, Sch Mechatron Engn & Automat, Shanghai Key Lab Power Stn Automat Technol, Shanghai, Peoples R China

年份：2018

卷号：159

起止页码：51

外文期刊名：KNOWLEDGE-BASED SYSTEMS

收录：SCI-EXPANDED(收录号：WOS:000446145600004)、、EI(收录号：20182405302470)、Scopus(收录号：2-s2.0-85048162362)、WOS

基金：This research is financially supported by China NSFC under grants 51607177, 61433012, U1435215, China Postdoctoral Science Foundation (2018M631005), and UK EPSRC grant under the Optimising Energy Management in Industry - OPTEMIN project EP/P004636/1.

语种：英文

外文关键词：Benchmarking - Embedded systems - Heuristic algorithms - Optimization

外文摘要：The majority of embedded systems are designed for specific applications, often associated with limited hardware resources in order to meet various and sometime conflicting requirements such as cost, speed, size and performance. Advanced intelligent heuristic optimization algorithms have been widely used in solving engineering problems. However, they might not be applicable to embedded systems, which often have extremely limited memory size. In this paper, a new compact teaching-learning based optimization method for solving global continuous problems is proposed, particularly aiming for neural network training in portable artificial intelligent (AI) devices. Comprehensive numerical experiments on benchmark problems and the training of two popular neural network systems verify that the new compact algorithm is capable of maintaining the high performance while the memory requirement is significantly reduced. It offers a promising tool for continuous optimization problems including the training of neural networks for intelligent embedded systems with limited memory resources.